Queens Bus Redesign (350159) | |
![]() |
|
Home > BusChat |
[ Post a New Response | Return to the Index ]
![]() |
![]() |
(350162) | |
Re: Queens Bus Redesign |
|
Posted by WMATAGMOAGH on Mon Jun 9 10:37:20 2025, in response to Queens Bus Redesign, posted by Stephen Bauman on Sat Jun 7 14:22:31 2025. Does your app work for any GTFS feed or only New York's? |
|
![]() |
(350163) | |
Re: Queens Bus Redesign |
|
Posted by Stephen Bauman on Mon Jun 9 13:11:34 2025, in response to Re: Queens Bus Redesign, posted by WMATAGMOAGH on Mon Jun 9 10:37:20 2025. It's still a work in progress. I developed a more meaningful metric than end-to-end bus speed. That's what I'm doing right now. I hope to finish it before the Redesign starts.That metric is pretty much NYC based: it's average travel time to/from census block and Manhattan CBD at 5 minute intervals throughout the 24 hour day. It includes bus and subway trip time from each of the 14,000 census blocks in Queens. The app can be expanded to other boroughs or combinations thereof. The bus speed results do not require NYC specific data. It does require the GTFS to have a shape file. That's optional and many operators do not include it in their GTFS. This is especially true for operators outside the US. There is an app to derive the shape file from the bus stop locations and trip stop times. It's proprietary from ArcGis. I'm working strictly from open source data using only open source applications. I tried to write my own but the results were not particularly good. I've discovered an open source application I did not try. I was about give it another try but the MTA released the GTFS. So, that try is on the back burner. I did develop a metric that does not require the shape file for comparing transit access. That's the walking distance between all the buildings and the nearest bus or subway stop. I calculate the percentiles for the walking distance to compare cities for the 800 m (1/2 mile) walking threshold to a subway entrance or 400 m threshold for walking distance to a bus stop. That's the extent of my excursions outside NYC. |
|
![]() |
(350164) | |
Re: Queens Bus Redesign |
|
Posted by WMATAGMOAGH on Mon Jun 9 13:26:47 2025, in response to Re: Queens Bus Redesign, posted by Stephen Bauman on Mon Jun 9 13:11:34 2025. WMATA just published their network redesign GTFS for its launch on June 29. I have the ArcGIS Personal Use License ($100/year) and started playing with the data yesterday. I have reason to believe there are some stealth cuts to service at certain locations that were not advertised at any point during the planning and outreach process. |
|
![]() |
(350165) | |
Re: Queens Bus Redesign |
|
Posted by Stephen Bauman on Mon Jun 9 14:46:22 2025, in response to Re: Queens Bus Redesign, posted by WMATAGMOAGH on Mon Jun 9 13:26:47 2025. I have reason to believe there are some stealth cuts to service at certain locations that were not advertised at any point during the planning and outreach process.I don't think you need anything as elaborate as what I'm doing to prove or disprove you suspicion. I'm using Python as the programming language because it has a vast library of packages to handle GTFS. Look into gtfs_kit, if you have not already. It will read the feed (as file or url) and spit out the various files as a pandas DataFrame or geopandas GeoDataFrame. Supply a date and it will go through all the gymnastics of going through the calendar and calendar_dates tables to the trips, routes, stop_times and stops tables for the specified date. It does have a bunch of functions that might answer your question directly. I've been just doing my own exercises on the DataFrame and GeoDataFrame tables for my work. Sounds simple, until you try it. :=) |
|
![]() |
(350166) | |
Re: Queens Bus Redesign |
|
Posted by WMATAGMOAGH on Tue Jun 10 09:03:05 2025, in response to Re: Queens Bus Redesign, posted by Stephen Bauman on Mon Jun 9 14:46:22 2025. This is far from my first foray into playing with GTFS. About 5 years ago, I was getting training in how to use R, but then I got laid off from that job and never picked it up on my own. However, even without R, I've created Marey Charts, block charts, and a whole suite of other analysis tools, mostly in Tableau. I've never used Python, and wasn't planning to start now. I was just curious if your work could be carried over elsewhere. |
|
![]() |